Tuesday, 15 April 2014

Week of 2014 AL 11

IWM fell by 3.5% this week, while my account fell 0.7%.  The loss-floor has fallen to -4.3%.  Only two more weeks before I can rescale this chart!

My account’s Y-T-D loss is now slightly smaller than the broad market’s loss, so I’m “beating the market” for now.

I didn’t feel like updating the following %gain chart on Monday.

End of week allocations:

33% short equities, 79% cash
% gain
size
Max loss
Sym  Buy  Fri  Mon  Tue  Wed  Thu  Fri Beg   End
 
ATML AL01 -0.2 14 -0.2 14 -0.2 14 -0.2 14 -0.2 0 -0.9 -0.9
TNA AL09 +0.1 10 -0.7 10 -0.7 10 -0.7 -0.7
URTY AL10 -0.4 13 -0.4 13 -0.8 -0.8
TZA AL10 -0.0 10 +0.4 11 -0.9 -0.9
IWM   -0.6    ?   -1.3   +0.1   -2.8   -4.1
me -2.3 ? -2.3 -2.3 -3.4 -3.0
floor -2.9 ? -2.3 -3.0 -4.3 -4.3

Stock-trading robot

The robot worked fine until Wednesday, then broke hard on Thursday.  Apparently the new version of Opera runs very close to my cloud server’s memory limit.  Something happened that caused it to consistently use slightly more memory and then nothing worked.  So I upgraded my server instance to have more memory, which will cost me extra money every month.  This is irksome because I’m sure there must be some way to cut the memory usage slightly, but I don’t feel lik putting in the research effort to find out what that way might be.

Ticker
Symbol
Buy dateBuy priceSell dateSell priceAcct Profit
ModelActualMAModelActual M  A  M  A 
ATML AL 01 AL 01 09:30 $8.34 AL 07 09:30 $8.23 -0.22% -0.23%
AL 04 11:45
SRTY ¹ AL 04 14:00 (Skipped) $41.34 AL 07 15:00 $43.19 +0.43% +0.00%
TNA AL 09 15:00 $74.36 AL 10 14:00 AL 10 13:38 $69.41 $69.36 -0.70% -0.72%
URTY AL 10 11:00 $82.32 $82.27 AL 10 14:00 AL 10 14:03 $80.16 $80.02 -0.37% -0.40%
SRTY ² AL 10 11:00 (Skipped) $42.48 (Not yet) +0.00%
TZA AL 10 15:00 $17.29 $17.30 (Not yet)

ATML: Loss is slightly larger than expected, probably because there were actually two fees for purchases while the model expected only one.

SRTY ¹: Skipped this trade because the robot was not working last week.  Would have been a winner (for a change)!

TNA: Worst possible timing — price fell every hour from purchase time to stop-out time.

URTY: Purchase went through, but then the robot ran out of memory while trying to create the protective-stop order, which I entered manually.  Hours later I entered the sell-order manually.

SRTY ²: Skipped this trade because the memory fault prevented the robot from buying it.  So far it looks like it would have been a winner.  It’s odd that this short model gave a “Buy” signal at the same time as the long URTY model.

TZA: So far, so good!

No comments:

Post a Comment